[sdiy] CA3080 switch driving voltage

[Grant Richter]

> 60mV - much lower than I expected.
> Doesn't this mean that there is a temperature drift from the diff pair
> still working in the "almost linear" area?
> But anyway - if saturation must be avoided, then this is the way to go.
> Probably compensating for the tempco somewhere else. (I remember
> there was another EN article with diodes setting the thresholds of
> the schmitt trigger - maybe this was exactly for that reason.)

Are you referring to this? From an RCA ap-note. Response is only 45 ns, with
1 ma. See:

http://www.musicsynthesizer.com/DIY/Grant/OTA.html

> 
> Thanks for pointing this out. All my previous triangle cores show
> the effect you describe in EN#112. Not tracking ore than 3 octaves
> accurately, and HFT trimmers won't help. This time I'll try the 60mV.

I'm using just a 100K in series with the NJM13700 input. It will zener at
around 7 volts. Works OK for an LFO, I'll have to see how it tracks as a
VCO.